Another Crazy Day in AI 34/104

Another Crazy Day in AI: Microsoft Bets on Agent Management

May 21
4 min read

Updated: May 22

Hello, AI Enthusiasts.

Midweek check-in: How’s your week going? The coffee’s cold, the AI news is hot.

While you're winding down, Microsoft's Satya Nadella is winding up the next era of work. The vision? Less grind, more guiding agents. It’s giving CEO energy... for everyone.

Also happening: U-M students are getting trained up in AI, and U.S. workplaces are seeing a sharp jump in generative AI adoption.

Ready for a quick scroll before calling it a day?

Here's another crazy day in AI:

The future of work according to Microsoft
University of Michigan offers free AI and career courses with Google
New survey shows AI use at work surged 43% in early 2025
Some AI tools to try out

TODAY'S FEATURED ITEM: Microsoft on Building the Agentic Web

A robotic scientist in a classic white coat with 'AI Scientist' on its back stands beside a human scientist with 'Human Scientist' on their coat, looking towards the AI Scientist.

Image Credit: Wowza (created with Ideogram)

What happens when most of the world’s code is written by AI?

That’s not a hypothetical—it’s where we’re headed, and Microsoft is leaning into it. In an exclusive interview at Build 2025, Rowan Cheung, founder of The Rundown AI, sat down with Microsoft CEO Satya Nadella to explore the future of work, the web, and our roles in an AI-powered world.

This featured conversation offers a first look at Microsoft’s vision for what Nadella calls the “agentic web”—a shift where AI agents aren’t just helpful tools but fundamental parts of daily workflows. From knowledge work to healthcare, Nadella makes the case that AI won’t replace us—but it will radically reshape how we work, what we build, and how fast we get there.

Here’s what’s taking shape in Microsoft’s AI strategy:

Microsoft envisions an “agentic web,” where intelligent agents handle complex, multi-step tasks across platforms.
The idea of a knowledge worker is evolving—future roles may focus on managing and coordinating AI agents.
AI is on track to generate the majority of code, but human oversight and decision-making remain essential.
Enterprises can now fine-tune Copilot tools with their own proprietary data, creating more useful and personalized AI outcomes.
Real-world applications, like tumor board meetings at Stanford Medicine, show AI’s potential to streamline healthcare processes.
Nadella emphasizes the need for cultural readiness, ongoing reskilling, and diffusion of tools across teams—not just in tech, but everywhere.

There’s a quiet but important shift happening here: AI is no longer framed as a singular tool or app, but as infrastructure. Microsoft’s approach isn’t just about product launches; it’s about rethinking how platforms, tools, data, and human decision-making come together. The emphasis on orchestration—of data, of agents, of intent—suggests that the future of productivity will be measured not only in speed or automation, but in how well systems work together with people.

For companies, the message is less about jumping on the latest hype and more about laying solid groundwork. It’s about experimenting with workflows, adapting team structures, and giving individuals room to engage with these tools meaningfully. As Nadella points out, successful adaptation won’t hinge on chasing trends—it will come from people within organizations who take initiative, spot opportunities, and build new ways of working from the inside out.

Watch the full interview here.

Listen on Apple Podcasts here.

Listen on Spotify here.

OTHER INTERESTING AI HIGHLIGHTS:

University of Michigan Offers Free AI and Career Courses with Google

/Lisa Gevelber, Founder, Grow with Google, on Google Blog – The Keyword

More than 66,000 University of Michigan students now have free access to Google’s AI and career training courses through a new partnership between the university and Google. The initiative includes credentials in high-demand areas like cybersecurity, UX design, data analytics, and access to new AI-focused courses like Google AI Essentials. Students can build job-ready skills, connect with an employer consortium of 150+ companies, and learn how to use generative AI tools responsibly. This collaboration expands on U-M’s ongoing work with Google and reinforces the growing importance of AI literacy across all disciplines.

Read more here.

New Survey Shows AI Use at Work Surged 43% in Early 2025

/Ethan Mollick, on X, referencing research by Jonathan S. Hartley, Filip Jolevski, Vitor Melo, and Brendan Moore, in The Labor Market Effects of Generative Artificial Intelligence

A nationally representative U.S. labor survey shows that generative AI use at work has jumped significantly—from 30% in December 2024 to over 43% by March/April 2025. Tools like Gemini and ChatGPT are seeing sharp increases in workplace adoption, particularly among younger, educated, and higher-income workers. The study analyzes how AI is changing productivity, efficiency, and job search behaviors across sectors. These findings point to a critical moment for businesses and policymakers as AI integration accelerates across the economy.

Read more here.

Hartley, Jonathan and Jolevski, Filip and Melo, Vitor and Moore, Brendan, The Labor Market Effects of Generative Artificial Intelligence (December 18, 2024). Available at SSRN: https://ssrn.com/abstract=5136877 or http://dx.doi.org/10.2139/ssrn.5136877

SOME AI TOOLS TO TRY OUT:

Clemta Intelligence – Automates bookkeeping, tax prep, and reports for businesses.
Distro – AI podcast host that turns episodes into posts, quotes, and drafts.
Wordtune – Rewrites your thoughts clearly and compellingly.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is on LinkedIn!!!

Leveraging AI for Enhanced Content: As part of our commitment to exploring new technologies, we used AI to help curate and refine our newsletters. This enriches our content and keeps us at the forefront of digital innovation, ensuring you stay informed with the latest trends and developments.

Another Crazy Day in AI: You Can Now Take NotebookLM Anywhere

May 20
4 min read

Updated: May 22

Hello, AI Enthusiasts.

Feeling done for the day? Here’s a quick one before you log off.

NotebookLM just went mobile, so you can turn heavy reading into easy chats wherever you are—subway, sofa, or between meetings.

Talent-wise, AI skills are now the hottest ticket in tech—51% of leaders say they’re hunting for talent like it’s a rare Pokémon. Gotta catch ‘em all, or at least upskill fast.

Oh, and some big papers accidentally served fake AI-written summer reads. Plot twist: editors didn’t see that coming.

Curious yet?

Here's another crazy day in AI:

NotebookLM gets a phone app
How to fix the AI skills shortage before it hurts your business
How fake AI book picks ended up in real newspapers
Some AI tools to try out

TODAY'S FEATURED ITEM: NotebookLM In Your Pocket

Image Credit: Wowza (created with Ideogram)

What if your go-to research assistant could now travel with you — even offline?

Google just launched the mobile version of its popular AI note-taking assistant, NotebookLM, and it’s packed with features designed to make understanding complex material easier—whether you're at your desk, on the subway, or in between meetings. In VentureBeat, executive editor Carl Franzen offers a closer look at the new app, unveiled during Google’s 2025 I/O conference. Initially a web-only tool, NotebookLM built a reputation for turning dense documents into conversational audio summaries. Now, all of that capability fits in your pocket.

Listen to Audio Overviews offline | Source: Google

Here’s what the latest version brings:

You can now access NotebookLM even without an internet connection, which makes it more flexible in real-world use.
Upload materials directly from your phone—PDFs, YouTube links, websites, and more.
Audio Overviews let you listen to summaries like a podcast, complete with background playback.
A new "Join" button lets you explore more context and dive deeper into topics with follow-up questions.
The Plus plan offers shared workspaces and deeper integrations geared toward teams.
Google confirms your files are not used to train AI models, maintaining user privacy.
Performance is still a work in progress, with some upload and responsiveness lags noted.
A Video Overview feature is in the works, promising animated summaries down the line.

Interact with the hosts | Source: Google

The move to mobile makes NotebookLM more flexible, especially for people constantly switching between devices or juggling work on the go. Instead of needing a quiet space to sit down and sift through material, users can now listen to content while commuting, walking, or taking a break between meetings. That shift—toward truly ambient learning—could open up new possibilities for how information is absorbed and revisited in the flow of daily life.

Share to NotebookLM from anywhere | Source: Google

Of course, it may not replace deeper research sessions at a desk or full document analysis on a large screen. But it does add another layer of interaction—especially helpful for educators, students, writers, or anyone who regularly works with complex ideas. With some thoughtful use, NotebookLM on mobile could become a quiet but steady support tool—ready when you need a second brain, even when your hands are full.

Read the full article here.

OTHER INTERESTING AI HIGHLIGHTS:

How to Fix the AI Skills Shortage Before it Hurts Your Business

/Mark Samuels, Senior Contributor, on ZDNET

Demand for AI skills has surged faster than any other tech area, with 51% of global tech leaders now reporting a shortage — nearly double from last year. To address this widening gap, businesses must rethink hiring, upskilling, and how they engage emerging talent. From investing in role-based certifications to sourcing AI-native professionals and reworking recruitment models, the solution is multi-faceted. Leaders who foster continuous learning and experimentation are more likely to attract and retain top AI talent.

Read more here.

How Fake AI Book Picks Ended Up In Real Newspapers

/Elizabeth Blair, Cultural Trends Correspondent, on NPR

A syndicated "Summer Reading List for 2025" mistakenly published by several major newspapers included fake AI-generated titles by real authors. Books attributed to writers like Isabel Allende and Percival Everett never existed, prompting backlash from readers and the literary community. The list, lacking a byline, was licensed through King Features, and writer Marco Buscaglia has since taken responsibility, admitting AI was used. The incident underscores growing tensions in publishing around AI-generated content and the erosion of editorial oversight.

Read more here.

SOME AI TOOLS TO TRY OUT:

Deepgram – Fast, accurate voice-to-text in seconds.
en.app – Find books by vibe, mood, or even vague descriptions.
Fliki – Create videos from text with AI voiceovers and clips, no editing skills needed.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is on LinkedIn!!!

Another Crazy Day in AI: Study Finds Smart Tools Fall Short on Basic Time Tasks

May 19
4 min read

Hello, AI Enthusiasts.

How’s your Monday night going? Maybe you’re still catching up from the weekend—or recovering from the first workday of the week. Either way, if you thought your brain felt a little foggy today, don’t worry: even the most powerful AI models are tripping over tasks we mastered in kindergarten.

A new study found that even top models struggle with reading analog clocks and figuring out calendar math. Turns out, some human skills are still surprisingly hard to replicate.

But for content creation, AI excels at idea generation and repurposing—if you know where to draw the line. Meanwhile, clinical AI is quietly improving patient care by assisting with scans and vitals.

Let’s see what the rest of the week holds.

Here's another crazy day in AI:

New study shows AI can’t reliably tell time
How to balance AI and human input in content strategy
What is clinical AI—and how is it used in care?
Some AI tools to try out

TODAY'S FEATURED ITEM: Why Smart AIs Still Can’t Tell Time

Image Credit: Wowza (created with Ideogram)

Can you tell time better than AI?

What seems like a basic skill for humans—telling time or figuring out what weekday a date falls on—still stumps even the most advanced AI models. A new study, featured at the ICLR 2025 Workshop and written by science journalist Drew Turney for Live Science, explores these surprising weaknesses in large multimodal AI systems. Based on research by Rohit Saxena and colleagues at the University of Edinburgh, the study tested how well leading AI models handle tasks like reading analog clocks and understanding calendar dates—and the results are eye-opening.

Despite excelling at language, image generation, and even passing professional exams, these models consistently fail at interpreting clocks and calendars—tasks most kids master by elementary school. The research points to key gaps in spatial reasoning and logical arithmetic—skills crucial for AI to operate in real-world, time-sensitive applications like scheduling, automation, and assistive tools.

Where the models struggled:

Clock reading accuracy was low across the board—even simple, standard clocks were misread.
Designs with Roman numerals or missing second hands made the task even harder for models to parse.
Calendar questions involving lesser-known dates or those requiring date calculation led to frequent errors.
One model (GPT-o1) performed relatively well on calendar tasks, but others showed inconsistent or near-random results.
Rather than computing answers, models often relied on pattern prediction from training data—leading to logic gaps.
Tasks that blend visual interpretation with structured reasoning continue to expose AI’s current limitations.

These findings offer a useful window into how today’s AI models process—and sometimes misprocess—visually grounded tasks that require structured logic. It's not just a matter of recognizing what's in an image, but being able to apply a sequence of reasoning steps to interpret it meaningfully. Reading a clock, for instance, involves understanding the spatial relationship between hands, accounting for visual variations, and then translating that into a precise time. Similarly, calendar reasoning often demands date arithmetic, which doesn’t come naturally to models that rely on predicting likely answers over calculating exact ones.

As AI becomes more embedded in everyday tools and workflows, understanding these limitations matters. Not to cast doubt on its capabilities, but to develop better expectations around what these systems can and can’t do yet—and where human oversight remains essential. This study underscores the importance of testing AI on the kinds of small, specific tasks that reflect how we actually use technology day to day. In that context, even something as ordinary as telling time becomes a meaningful benchmark.

Read the full article here.

Read the full paper here.

OTHER INTERESTING AI HIGHLIGHTS:

How to Balance AI and Human Input in Content Strategy

/Adam Tanguay, Head of Growth at Jordan Digital Marketing, on Search Engine Land

AI can speed up content creation, but it can’t replace human creativity or strategic thinking. Adam Tanguay outlines a scalable, client-ready SEO content process that uses AI where it excels—ideation, structuring, and repurposing—while keeping humans in charge of originality, accuracy, and brand voice. His agency builds a custom GPT per client, feeding it brand-specific data to streamline workflows and scale content without sacrificing quality. The result: more efficient teams, better SEO, and content that still feels human.

Read more here.

What Is Clinical AI—And How Is It Used In Care?

/Madeleine Streets, Senior Content Manager, on TechTarget

Clinical AI refers to AI systems used directly in medical care—like helping doctors interpret scans, monitor vitals, and support treatment decisions. Unlike broader healthcare AI, which includes billing or marketing applications, clinical AI is focused on improving patient outcomes in settings like hospitals and urgent care. It augments the work of doctors and nurses by spotting early signs of illness, tracking patient data in real time, and automating documentation. While still evolving, clinical AI is becoming a crucial tool in hands-on patient care.

Read more here.

10 applications of AI in healthcare

SOME AI TOOLS TO TRY OUT:

Inkr – Converts recordings into searchable transcripts and AI-powered notes.
TaoPrompt – Quickly craft professional AI prompts with accuracy.
Pinch – Real-time voice translation in 30+ languages for video calls.

That’s a wrap on today’s Almost Daily craziness.

Catch us almost every day—almost! 😉

EXCITING NEWS:

The Another Crazy Day in AI newsletter is on LinkedIn!!!